Estimation of human arm impedance using a nonlinear musculoskeletal model for posture and movement control

Understanding human arm impedance is crucial for applications in robotics, human-machine interaction, and rehabilitation. This study presents a musculoskeletal model of the human arm to estimate endpoint impedance during posture maintenance and movement. The model incorporates eight muscles (six monoarticular and two biarticular) and employs the nonlinear Hunt-Crossley model to capture muscle viscoelasticity. Joint torque and muscle forces are computed to derive endpoint stiffness, viscosity, and inertia, represented as ellipses in Cartesian space. Simulations compare the proposed Muscle Space Model (MSM) with a conventional Joint Space Model (JSM). Experimental validation confirms that the MSM better approximates human arm impedance, with stiffness increasing by up to 40% compared to the JSM. Results demonstrate that muscle redundancy significantly influences impedance modulation, particularly in dynamic tasks. These findings provide insights for designing adaptive robotic controllers and prosthetics that mimic human biomechanics.